Amplitude is now a $1.5B public company. Here's how they beat competitors with a 10x cheaper product. | Jeffrey Wang, Co-Founder of Amplitude

When Amplitude launched Mixpanel was the big game in town. They were first to market, had raised more money, and had a well-known brand. VCs passed on Amplitude because it seemed like just another Mixpanel.

Today, Amplitude is a $1.5B public company—they're about 2x bigger than Mixpanel. Mixpanel's marketing spend helped educate the market. But before buying an analytics solution most businesses do market research. That's when they'd find out that Amplitude had several features Mixpanel lacked—and they were much, MUCH cheaper.

It's not cool to win on price, but it works. It worked for WalMart, CostCo, Shein, and it worked for Ampltiude.

Here's the story of how it all happened.

Why you should listen:

• How to use cheaper prices to win in a crowded market.
• Why you often need 12 hour days to win in Startupland.
• Why even massive $1B+ successes often have trouble raising early rounds.
• How pivoting can often be the key to finding real market pull.
• Why big competitors can often be a huge tailwind.
• How to use storytelling to raising bigger rounds. 

Keywords
startups, entrepreneurship, analytics, Amplitude, pricing strategy, market positioning, data processing, voice recognition, technology pivot, competitive advantage, market dynamics, differentiation, product-market fit, storytelling, fundraising, startup challenges, customer relationships, analytics tools, business strategy, entrepreneurship

Timestamps:

(00:00:00) Intro
(00:06:13) A cool demo-- but a bad business
(00:18:36) Why funding was so hard
(00:25:43) Why lower prices are a big differentiator
(00:40:50) Working 24/7
(00:50:35) Product Market Fit

Chapters

• 6:13 A cool demo-- but a bad business
• 18:36 Why funding was so hard
• 25:43 Why lower prices are a big differentiator
• 40:50 Working 24/7
• 50:35 Product Market Fit

Transcript

Jeffrey Wang (00:00):

If you can make it a qualitatively different price profile. I think that's absolutely as much of a competitive advantage as product quality or anything like that. We came back from the winter break and New Years and everything, and then on January 4th we accidentally ran a script that deleted a bunch of critical databases.

Pablo Srugo (00:21):

Oh man.

Jeffrey Wang (00:22):

And this was at 10:00 PM we were in the office, someone accidentally ran a script and I was like, oh shit. It's like, wait, what do we do? It's like this thing's gone. Everything, all the alarms are firing because the whole site's down. Our whole data collection pipeline is down these critical databases. And we're like, oh shit. It's like we're panicking, we're calling up AWS. It's like, Hey, can you recover this? We accidentally deleted these three databases. 

There are some domains, marketing, fundraising, where I think you get disproportionate leverage out of your ability to story tell and exaggerate in ways that, not trying to lie, but trying to communicate your point again in a sharper and more emotional way to people. We lovingly refer to it now as the China bailout of Q1 2016. We actually signed, believe it or not, we signed ByteDance as a customer in 2016 for $20,000 a month. 

Previous guests (1:20)

That's product market fit, product market fit, product market fit. I call that the product market fit Question, product market fit, product market fit, product market fit. product market fit. I mean the name of the show is Product Market Fit.

Pablo Srugo (01:32):

Do you think the product Market Fit show has product market fit? If you do, then there's something you just have to do. You have to take out your phone, you have to leave the show five stars. It lets us reach more founders and it lets us get better guests. Thank you. 

Well, Jeff, welcome to the show, man.

Jeffrey Wang (01:47):

Thanks, Pablo. Glad to be here. Thank you for having me.

Pablo Srugo (01:49):

So I mean Amplitude is now a massive company, public company worth well over a billion dollars. And you were there effectively since day one. I mean, your co-founder joined a little bit after everything started, but still effectively pre-revenue or very, very early revenue certainly before things absolutely-

Jeffrey Wang (02:10):

That's right. That's right.

Pablo Srugo (02:11):

Would love to dive back to you in those days. Maybe just tell us just a little bit of your background and how you came to be part of this equation and work with the early co-founders of Amplitude.

Jeffrey Wang (02:23):

For sure. For sure. Yeah, so my background, so I studied computer science at school, nothing too crazy. Having gone to Stanford and being in the Bay Area, you can kind of see the whole startup environment and especially in the early 2010s. It was really hot and that was the thing that everyone was thinking about. One of the things I always knew that I wanted to do was get involved in some way, but I was a little less bold than my two co-founders, Spencer and Curtis. And so I spent a few years working at small-ish companies, worked at Palantir for a year.

Pablo Srugo (02:55):

What made Startup so hot back then was this, I guess Facebook had just gone public. Were these some of the big things in people's minds?

Jeffrey Wang (02:59):

Yeah, so back in that time, I think a lot of companies were starting around that area and Facebook in particular was a big deal. And so there was all this energy around entrepreneurship and startups and that being the pull of really great TS students at the time. And so that was just the buzzy thing. There were entrepreneurship kind of clubs and communities coming around at the campus and just in the area, and that's what people were talking about.

Pablo Srugo (03:24):

But you went off a different track. So you went off to Palantir and I think Sumo Logic or something like that.

Jeffrey Wang (03:29):

Yeah, so at the time Facebook was the one for people who thought startups were cool, but didn't want to start one. Palantir was number two on that list. Palantir, for what it's worth, it was pretty small at the time, like 300 people. But they did a really, really good job of building this culture of the best engineers. And one of my funny sayings is I think Palantir's greatest competitive advantage was the fact that they managed to convince really, really smart people to work on enterprise Java applications and that was their biggest competitive advantage. 

Pablo Srugo (04:02):

Even now I hear Palantir and I'm like, Palantir man. You know what I mean? There's something about it.

Jeffrey Wang (04:06):

There was the secrecy and then like, oh, we're like all working on this thing- Inn practice, it was just a normal software company. And so I went there and it's like, yeah, I did some table applications and we ship software on CDs to the government. There was nothing glamorous about it, but the people were cool.

Pablo Srugo (04:23):

And so how does that kind of transition happen?

Jeffrey Wang (04:27):

Yeah, and so I worked at Palantir, Sumo Logic and what I realised, so I did one more, which was an actual small startup. It was a 10% company called Clinical. They're a little bit infamous if you google some of the news articles from back in the day, but I won't go into that one. But what I realised, and that was less than 10 people, so what I realised was as I got into smaller and smaller companies, the more kind of excited and invested and just motivated I felt about the work I was doing. And in particular, I got to work on some pretty interesting and big large scale data processing problems, which were kind of in the domain of interest. And so eventually I was working for these companies and I was kind of thinking about it. I was like, oh, okay, I'm learning stuff and I'm growing and it's pretty interesting, but there's something a little bit missing.

(05:12):

I wanted to just throw myself at something that I really believed in with people that I really wanted to work with and I was kind of passively looking for that. And so a year and a half into working at Sumo Logic, my co-founders, Spencer and Curtis, they reached out to me. They had to- to go into their story a little bit. They had gone through YC. They were actually doing a text-by-voice company called SonaLight. And so they were trying to do Siri before Siri is really a thing, and they were doing it for Android and going into that story a little bit more, but they've eventually pivoted away from that to do analytics and that kind of matched with my interests with data processing and all that. And we happened to meet through some mutual friends and they were like, Hey, we're looking for more good engineers to come and build, solve this hard technical problem and build some really cool software. And so we hit it off after a couple of meetings and then I eventually took the leap.

Pablo Srugo (06:07):

And had they already shifted to the Amplitude play or were they still on the previous product?

Jeffrey Wang (06:13):

Yeah, so they had already shifted. So to give a little bit more context on the story, they had gone through YC. They did Demo Day with their old product Text by Voice.

Pablo Srugo (06:21):

What year is this that they went through YC?

Jeffrey Wang (06:23):

This was winter ‘12, 2012, and their demo day was actually incredible. I think they had the most hype of their batch and all these people were like, this is the future of text-by-voice and how we going to control our phones- it was,

Pablo Srugo (06:37):

It's funny. Demo day is very, I mean it's “the medium is the message” sort of thing. It's one specific thing. It's 3-10 minutes on stage, boom. If you've got some back office thing, it could be a great business, but it's so boring. 

Jeffrey Wang (06:51):

yeah and it’s so hard of the pitch,

Pablo Srugo (06:52):

But if you've got- the Siri, that's just hype. We did DoubleDay Power Stars with Gym Track and it was about tracking your workouts. We had this video of people working out all these analytics that got written about because it was cool. business completely flopped, but it was a great demo.

Jeffrey Wang (07:05):

Yeah, yeah. It's certainly an interesting dynamic and I think there's still a YouTube video of their presentation. Spencer walked up onto stage and he was just talking and then his phone was in his pocket and just picking up what he was saying and it was texting people and was like, yeah, to your point, it's just kind of this perfect demo environment for that type of product. And so they did demo day, it was super popular. They finished YC and then they raised their seed round, a couple million bucks off of that. It was at that same time that they realised, Hey, we don't think we can turn this into a long-term business. And I'll go into a little bit of the details of that, but basically they had spent a bunch of the time looking at their usage and understanding how their users were using the product, what was good, what was bad, kind of your standard introspective, analytical.

Pablo Srugo (07:52):

Yeah. Tell me a bit more. what was it, I think it's Siri, but it's kind of Siri, almost like an interface for other apps. What was the standalone app text primarily?

Jeffrey Wang (08:01):

One: read your text to you? And that's pretty straightforward, but then two: voice recognition, so you could send texts without touching your phone. This is because you want to be able to drive and text and do other things while being able to text. And again, a really useful and practical idea in practice. And what they realised was it wasn't that the idea was bad, it was that through their analytics, they saw that if users encountered errors or basically mistranslations or whatever of their speech to text, they were only a little bit tolerant of that. So the first time it fails, that's cool. People were willing to give it a second chance, second time it fails, they're gone. And so that was the amount of tolerance that they had. And so what they realised was that, hey, to actually make this product successful, we need to dramatically improve the voice recognition technology. And they're not machine learning researchers. In fact, they were just using the Google Voice API to do the speech recognition, and so they could have spent the next five to 10 years kind of banging their heads against voice recognition as a problem, but they kind of realised that that wasn't their strong suit, and so they started looking into other options.

Pablo Srugo (09:13):

It's an interesting, I'll say something just a tangent on there because it reminds me again of my story a little bit, which is at least in consumer, I don't know if B2B is actually like this, but consumer is very much zero to one in the sense of it either works or doesn't work. The reality is it's probabilities. Let's say it was 80% accuracy or 85% accuracy, but you almost hit this point where there's enough accuracy that people think it works and below that line it doesn't work. And literally that's how people think about things. Same thing with GymTrack. GymTrack was about counting reps and it was- you would think if I said to you, “this thing's 90% accurate”, that's pretty good. That sounds like a good accuracy, but if I tell you you did nine reps when you did 10, it's absolute garbage. It just doesn't work. And I think for some things, I don't know how to make sense of this, but I always compared it to running. A GPS. You don't really know. So if something tells you you ran five kilometres, you're like, cool, I ran five kilometres. You later find out it was 4.89. Unless you're an elite runner, you're like, wow, right. How else am I going to measure? It seems to work pretty well, but it's kind of similar to that. You think about text-to-speech, you can do it yourself perfectly. So if the machine's going to do it for you, it needs to be pretty much perfect. Otherwise you're like, it just doesn't work, it's not useful. And then the value drops to zero pretty quick.

Jeffrey Wang (10:24):

Yeah, I think that model of thinking is right for a lot of domains where there's kind of these natural attraction points, if you will, of constant usage or zero usage, and then depending on the accuracy, you're either enough to get to that attraction point of a hundred percent usage where people are just depending on this all the time or you're not high enough and then you just immediately get dragged to zero. That's right. I think there are actually some interesting analogies between that and the current trend of AI and LLMs, but that's a different topic.

Pablo Srugo (10:51):

No, I mean there's a lot of parts to it. At the beginning it just felt like it was going to take over everything, and maybe it will, it'll take more time, but there's definitely some people who've tried for certain things and it didn't work and now it's done. It just doesn't work. But the stuff evolves kind of in a line. I think if you're a consumer, it just is a step function, but in the background it's kind of a bit more of a line.

Jeffrey Wang (11:14):

Yeah, yeah, definitely. Interesting. And so yeah, that realization was very important for Spencer and Curtis to make in those early days because it kind of dictated the next 10 years of their life, if you will. But the way the tool that they used to do that was this internal analytics thing that they had built because it is funny because now looking back, they're like, oh, wow.  It doesn't make sense if you're trying to build a successful consumer product to spend a bunch of your time building this internal analytics tool, it's not actually helping you do the thing. But at the same time, it also helped them discover the thing that they wanted to pivot to because once they made that decision, all these other companies in their YC batch were like, Hey, that tool that you showed us that you used to make your decision to pivot your company, that seems actually pretty cool.

(12:00):

We could use that for our own products and discover all these interesting insights about how our users behave and what matters to 'em and what doesn't. And so the combination of those two things happening at the same time made them think, Hey, maybe we should just pivot to this analytics thing that we built. They had tried a bunch of other tools out there and they were fine, but they didn't actually answer the key question that they wanted to answer, which was: How does this particular behaviour, seeing errors in your speech recognition, how does that impact people's retention and how much they use it longterm? So a little bit more sophisticated of a question than what analytics products did at the time, which is mostly like, Hey, how many people came to my site? How many people click on this button? What countries are they from? What browser are they using? All useful stuff, but a little bit of a different type of question. And so that kind of type of thinking led them down the path of Amplitude.

Pablo Srugo (12:48):

It's interesting. I mean it reminds me very much of the Shopify story, which is kind of a classic Toby building a website to sell snowboards and he sold some snowboards, but people were like, man, that's a pretty sweet website. What did you use? And then he is like, okay, then Shopify. So it's funny how things come to be. When did you join? This is winter 2013. This is happening, when do you come into the picture?

Jeffrey Wang (13:11):

That's right, that's right. I think they officially pivoted in the summer of 2012, if I remember correctly. So they had done winter ‘12, so earlier that year. Then they pivoted right after I came in at the beginning of 2014, so actually a decent amount of time later. And they liked to joke- Spencer and Curtis liked to joke that they were kind of messing around during that time and couldn't figure out what was going on, but I'll give them a lot of credit. They made a lot of progress. They sold their first customer in late 2013, which was a gaming startup that they were close to, and so they had real revenue in the books. I want to say when I joined there was between 10 and 20 K of ARR and so not zero, not a lot, but not zero. And so they made a lot of progress during that time and a lot of it is like, Hey, talk to customers, understand what the pain actually is, and they always try to discredit that time, but I put a lot of weight into it where it's like,

Pablo Srugo (14:10):

Well, yeah, I'm curious. What did you see of the negative view, let's say would be, okay, so if you made the discovery, let's call it Q2 or so of 2012, it's 18 months later, certainly over a year later, and you've got a couple customers to show for it for something that it's not about simple or complicated product, but it's not like you're trying to go to Mars, you're trying to do analytics, but for mobile let's say or whatever, analytics is there for the web, and you’re trying to do it for mobile. Why is it taking so long? What was happening?

Jeffrey Wang (14:42):

Yeah, yeah, that's a good question. I think 1. There's a lot of iteration on the infrastructure, the shape of the problem. There were existing tools, but obviously there was something unsatisfying about them, which is why they wanted to build something new. And so understanding what that dissatisfaction was and how to translate that into product. And then the analytics more so than your traditional SaaS company or traditional consumer company has quite a bit of infrastructure/software challenges to it. Even at the time when I joined up with them, even though there were very few customers, very little revenue, they were actually doing quite a bit of scale, a very unnatural amount of scale relative to how much revenue they had. So they had to solve some pretty interesting technical problems even to get to that point

Pablo Srugo (15:29):

Because those customers just, they had a lot of users

Jeffrey Wang (15:32):

And customers that have a lot of users and a lot of data feel the pain a lot more. You got a database of a hundred people and it's like your analytics can be literally anything. You could use Excel and you're fine. And so it's like you don't have a lot of interesting problems with respect to this. And so only when you have billions of data points that you're like, okay, I got to figure out something. I can't just run my normal processes. I definitely can't use Excel. I can't even use my normal database to do this anymore because the queries are too slow. It'll actually take down my production site. And so it's really at that scale that it gets interesting as a kind of unique problem. And so they were actually processing quite a bit of data and I had worked at a few analytics companies, seen some large scale data processing, and actually they had worked that Spencer had done trading.

(16:22):

Curtis was doing Google photo stuff, and so they weren't really kind of familiar with how to solve problems in that way, and so they kind of had to learn that on the flight, which is incredible to begin with. But they were able to handle some really interesting scale by the time I got there. And so that was kind of what I saw. I was like, okay, these people seem smart, these companies sending them a lot of data to process that was kind of intrinsically exciting from a problem perspective. It's like, oh, I kind of want to go work on this. And the second thing I saw, which this may be an unsatisfying answer to a lot of people, but it's the real answer, which is I saw Curtis and Spencer's people and I spent time with them and I was like, Hey, these people, 1. they're incredibly smart, they're incredibly high integrity. They care a lot about this problem that they're working on and the fact that they want to throw themselves at this problem the same way that I was looking for a problem to throw myself at. And so that just kind of matching of personalities and where we were in our stages of life and what we cared about, it took me almost no time to become good friends with them and then be willing to just go hard for an indefinite period of time at something that had a small chance of success.

Pablo Srugo (17:36):

How many people when you joined it was just them two?

Jeffrey Wang (17:40):

They brought on two other people very recently at that point, but it was really, both of those people had been there for a couple months at that point, and so it was very early still.

Pablo Srugo (17:51):

How many people are in Amplitude today?

Jeffrey Wang (17:53):

Today we have 700, 750 or so. It's funny that you say we're really big and stuff. I still like to think of ourselves as small and early.  One: because that's just the vibe I like and two: there’s so much more to do. And so I definitely don't want to get complacent or anything. It's like, okay. Yeah,

Pablo Srugo (18:12):

There's always someone bigger. The image that sticks with me there, there's an interview of things like Mark Cuban and he was like, man, Warren Buffett is so rich. And I'm like, really? you’re still- you know what I mean, you’re still looking at people like that? so I dunno. Yeah, you can always look up, but you guys are big man. You guys are big.

Speaker 2 (18:31):

Yeah, 

Pablo Srugo (18:31):

The other question I had is, when you joined, I mean they raised that seed round a bit ago. What was the funding kind of situation at that point?

Jeffrey Wang (18:39):

So they were running off that seed round. It was like your true standard 2 million seed round back in the day, and so that was enough to last them. They were taking low salaries and traditional founder-

Pablo Srugo (18:50):

They still had a year of runway or so when you came?

Jeffrey Wang (18:52):

Yeah yeah, I  think probably more than a year of runway when I got there. Probably still two is my guess. They were running very lean, which is something that Amplitude has always been all about, is it being very lean and efficient in the way we build? And so it wasn't until, so we spent that year trying to get traction, can we get to the point where we haven't have traction to raise an A and we weren't really at risk of running out of money at the end of the year, but it would by the end of the year, those 2014 at the time, it would've gotten to the point where if we didn't have traction at the time, then you'd be like, oh, okay, now we're digging into the lifeline at the end and need to start seriously thinking about what the options are. But fortunately, we were able to drive a bunch of traction that year and so then we could raise that series A at the end.

Pablo Srugo (19:40):

And so that year- give me a bit more of the context. Where was Mixpanel at? Who else was there and maybe more related is when you're selling into these companies, are they analysing different solutions and picking amplitude versus or is it just complete lotion, there's just nobody there servicing that particular use case?

Jeffrey Wang (20:00):

Yeah, yeah, that's a good question. So there were a good number of competitors at the time at different stages and of different forms, and so you had Mixpanel, which had been around for a few years, which was the most direct competitor, and they were pretty popular in the startups that we were selling to at the time. The type of people that we actually had access to and could find and our big differentiator so to speak. We had two differentiates against 'em. One was just price. They were very expensive at the time, and you always feel a little bit weird about fighting on price, but in the world of analytics, it's actually a really interesting dimension because, talk to the vast majority of people, almost a huge percentage of them, the most worrisome thing about analytics for them is price. Because of this, they have a lot of data that's just the norm these days. They have a lot of data and they want to do interesting stuff with it, and price always gets in the way. And so it's almost a very fundamental part of analytics as a market and not really just this dimension that lets you get away with stuff and throw away things for free. It's actually fundamental and price relates to infrastructure and how efficiently you're building your infrastructure and how you can process data. And so it's very important-

Pablo Srugo (21:15):

That's what I was going to ask. Were just charging less or you had an actual cost advantage or believed you have at a cost advantage because of how you-? 

Jeffrey Wang (21:23):

I definitely believe we had a cost advantage. So I remember we did some analysis of Mixpanel's list prices at the time versus our internal costs, and the difference was a factor of a hundred. And  how much of it they've been taking margin, we don't know for sure, but we looked at that and we're like, it seems like there's an opportunity to disrupt this a little bit. I remember it used to be makes, if you put their little badge on your website, you would get 25,000 events for free. It's like, okay, 25,000 events. Well, if you're running a website, as soon as you start using that, you have, oh, what? It's already gone. Boom, there it is. And so when we came to market, and this is somewhat related to pricing, but actually related to another point around marketing, the biggest marketing success that we had for the first five or six years of Amplitude was we offered 10 million events for free to everyone. And so you can kind of imagine that actually disrupts it quite a bit where there were lots of customers of Mixpanel paying quite a bit of money for a lot less than 10 million events, and so that would immediately make them come look at it.

Pablo Srugo (22:31):

Walk me through that story because I understand price and I understand the idea of they're doing 25,000, let's beat them at their own game, a hundred thousand, even a million, but 10 million sounds very strategic. How do you get there?

Jeffrey Wang (22:43):

Yeah, yeah. Again, we kind of looked at our cost profile and we used that to kind of say, how far can we push the boundary on this? That still makes sense for this. Again, we're not trying to just throw away money, we're trying to make a decision that actually is sound from a business perspective. So 10 million events fully utilised at the time that would've cost us. There's 10 million events a month. To be clear, that would've cost us $20 a month for a fully utilised account. One, you have the fact that most of the time-

Pablo Srugo (23:17):

so they were being cheap. I mean they're giving away cents like here's 2 cents if you sign up.

Jeffrey Wang (23:21):

Yeah, it depends on how you're building your infrastructure.

I think there's something to appreciate about that. And Curtis and Spencer had done a good job of that. I came in and I kind of used a bunch of my background working on these app. Palantir and Sumo Logic were very, very high scale even though they were startups, but the nature of their businesses and being in analytics and data, they were dealing much, much higher scales. And so they learned a lot of the techniques for being efficient, and I was able to learn that from my time there, and that's why I really appreciate spending time at those companies. And so I was able to apply a bunch of those things at Amplitude and make sure that, one, we could have this nice cost profile and then two that if we suddenly started getting 10, 100 times as much data that we'd actually be able to support that. So the combination of those things made us very confident going into this 10 million events for free. And so yeah, you got $20 a month, most people aren't going to use those 10 million events because 10 million is actually quite a bit now you got some random personal website, you're not going to hit that probably ever. I mean in most cases, and so a couple dollars a month to get these accounts and we're B2B not consumers. And so you're not going to

Pablo Srugo (24:27):

that's an absolute no brainer, plus you get the positioning advantage, the marketing advantage on top of it all.

Jeffrey Wang (24:32):

Every single mixed panel customer would look at Amplitude at that point. And that was how we got off to our crazy start.

Pablo Srugo (24:38):

And they were the hotter one at that stage. They'd raised 10 million from A16. They were not really a big dog because it's not like they're massive but-

Jeffrey Wang (24:46):

They had reason behind them and everything, and so they were a very big deal at the time. And so there was a price thing and that was a big deal. There was also when I mentioned earlier around the type of analysis, if you will, Spencer and Curtis, when they were building tunnel light, they really wanted to do this. How does X behaviour seeing errors in your speech recognition affect Y behaviour where it's like retention, engagement, this kind of behaviour impacting behaviour analysis. And today, if you think about it and think about any analytics, well, they can pretty much all do it.

Pablo Srugo (25:18):

Yeah, it's all events driven. How many other people who do this event, what's the-

Jeffrey Wang (25:22):

It was actually a little bit unique at the time. There wasn't quite any other school that could do it in that way. And so we built that in alongside with the cost efficient infrastructure and the combination of those. So it's like, okay, people look at us because of the cost and they're like, oh, wow, you can actually do more analysis than before. It's like, wow, it's kind of a no brainer at that point.

Pablo Srugo (25:43):

What do you think? So there's a whole tangent here because, well, let's separate it. First of all, just price. There's such a stigma. I think in startups, in traditional business, price it’s understood to be a big deal.

Walmart won because of price. A lot of these big companies win just, they're just like, Hey, we're going to do it cheaper and then everybody gets it. But in startup land, it's a bit of a stigma. It's like, yeah, you want to win 10 x better, you want to win, you're easier to use because the UI is better, whatever, you don't want to win cheaper. But the reality is that's a good strategy. I mean, you see the big guys, you could dumb down the teams versus Slack thing to just Teams is basically baked in. It's like quote unquote free and they kind of win because of that. Not super clean, but a little bit, and you certainly, I've seen it, this is just top of mind for me, but a little bit on the side, there's this company in Toronto called Host Away in the Get in their software, think about property management's software for Airbnbs and these sort of things. Anyways, there's this big dog called Guestie. They're like a billion dollar company host away. All they did, and I asked people in the industry, all they did was they just matched the features dated for 20% the price. Now they're worth a billion dollars. I'm sure there's other things, but literally I asked founders in that ecosystem and I'm like, were they better in any way? They're like, no, no. They were just cheaper. And at some point people were like, why am I paying five times more for something that I could do? This is ridiculous. So all of that to say, because you saw it firsthand at Amplitude, you did an amplitude, maybe it's something you thought about more. What do you think about that? You think that there isn't enough focus on just trying to deliver something cheaper than whatever the incumbent or market leader is as a winning strategy?

Jeffrey Wang (27:27):

I think it depends a lot on the domain, but I would say that if you can make it a qualitatively different price profile, I think that's absolutely as much of a competitive advantage as product quality or anything like that. And so in our case, we kind of saw it as, again, the 100x difference. We're like, Hey, what's going on here? There's something that we can uncover that the market has not discovered in terms of efficiency perspective. And so in that case, it was pretty clear to us that there was something there. Now for other domains, I can kind of understand in some cases where-  There's a couple things going on. One, it's like, okay, being 20, 30% cheaper, that's not enough to really get people to think about it. And then two, I think there's the flip side of your argument, which is if there is not actually a price differentiation angle, there isn't actually one there, but now you try to make one, you kind of just dig yourself into a little bit of a hole where you're like, okay, we got all this traction because of price.

It's like, well now let's go make a sustainable business. And it's like, well, okay, that's not possible under the price profile that you kind of discovered it could still work, right? I mean, sure, there's examples. Maybe you look at stuff like Uber where it's like, okay, they didn't really have that much of an interesting price advantage, but they kind of throw a tonne of easy money at it. They established market dominance, then you have network effects. And so it could still work, but I think in a lot of other businesses trying to do that might actually turn into more problematic than you’d expect. And so in contrast, we believe that we had a qualitative and scalable price efficiency to figure out, and I think you see similar things around the LLM and APIs today as another example, it's kind of a similar model to us. We charge based on events, they charge based on API calls and it's like having a price advantage there is massive, right? It makes a really big difference. So the people using those APIs and the fact that if you actually come up with a lower powered smaller model that delivers high performance, that is a huge competitive advantage. I think people do understand it really well in a domain like that. So yeah, I think it depends on the type of problem that you're looking at and how much it comes to people's minds.

Pablo Srugo (29:44):

I think in the LLM side, people are starting to get the idea that a lot of this stuff is a bit of a commodity, at least in the sense of you put in a prompt, you get some sort of an answer back, some are better, one thing, some better others, but they tend to all kind of conversion, so then it just becomes about, okay, who's going to do a faster cheaper? Those are the two things that end up being top of mind. So I totally get that.

Jeffrey Wang (30:02):

Your point about it being a commodity is actually interesting because we definitely wouldn't have said it at the time, but to some extent we believe that the analytics infrastructure is a commodity now. And so what is our strategy? Our strategy is we're building a bunch of tooling around that analytics infrastructure because your end goal isn't to do analytics. Your end goal is to make good decisions and actually make your product better and power your product in interesting ways. And so in some sense, maybe that insight is the right thing. The fact that it was a commodity means that people are price sensitive about it, and so you can apply that to lots of domains and which part is the commodity and which part is the differentiation, the part that's the commodity. You basically don't expect to make much margin off of that, and so then you drive the price down as much as you can, and then you have all these other use cases which are not a commodity, and then you make your money off of that. And I think that's kind of standard business as well. 

Pablo Srugo (30:53):

Ihink that makes total sense. I mean in this particular example, and I think it's something maybe let's say hypothesis or whatever, but a lot of times what happens is the market leader moves up-market and their pricing is a little bit based on that and those types of customers and the feature set becomes very complicated because of those types of customers. That's this host away example. Then a host away looks at that like, okay, I'm not going to do this upmarket stuff. I'll charge way less. You could continue to charge way less. I don't have a cost advantage, but you're kind of shooting yourself in the foot more than I'm starting from nothing and you're starting from these big enterprise contracts and B, I'm not going to build the 20% that gets you those things. I'll build the 80% that everybody else needs and that can be a play too. But just to keep going on the Amplitude example, talk a little bit more about Mixpanels specific on the side of, given that, did you find that the fact that they existed was a benefit in the sense that they were maybe educating the market and spending a lot of marketing dollars? Was that happening for you guys?

Jeffrey Wang (31:54):

For sure, for sure. It definitely a benefit for us in that we didn't have to evangelise really at all at the beginning. Maybe we evangelise a little bit of our approach, but by and large people kind of understood this space. The fact that, hey, it's really useful to have an analytics tool that's tracking what's going on in your product so you can tell and make good decisions about your product. That was huge, and so the vast majority of our leads in the early days came from existing Mixpanel customers, and so the vast majority-

Pablo Srugo (32:23):

You love this show, you don't want to miss the next episode. Why would you? so hit that follow button? Trust me, it's in your own best interest.

 That's crazy. That's another kind of false stigma. It's like you want back the market leader, you do. You want to back a company that ends up being the market leader, which sometimes is the one that starts off as a market leader and many times is not that one. It's just in this case, you guys are now bigger and it's partially the market leader has to go and be a pioneer, has to go and evangelise and has to spend those marketing dollars. They tend to build a brand as a result, but a lot of B2B customers especially don't buy without some sort of a bit of a bake off at least like, oh, I'm going to do analytics, especially analytics. I'm not going to want to switch analytics every three months. Let's see what else is out there, and then it's kind of free leads for you guys.

Jeffrey Wang (33:11):

Yeah. One thing that I maybe wish we had done better, although it's not totally clear how when we pitched the VCs, especially in the early days, often the question or the response would be, well, hey, there's all these other analyst companies already.

Pablo Srugo (33:27):

Have you heard Mixpanel? <laughs>

Jeffrey Wang (33:27):

Why should we fund you? oh, yeah, yeah, we're familiar with them and we actually didn't have a great answer to that. We had some intuition about the types of problems that we're trying to solve, but it was a little bit too nuanced. It was like, okay, to a sophisticated analytics oriented customer, they could understand it, but for most people they actually couldn't understand the nuance, and so that actually made fundraising early. A challenge, even though we were getting revenue and share traction was like people still couldn't understand why, if that makes sense. It's like why are-

Pablo Srugo (33:58):

a hundred percent,

Jeffrey Wang (34:00):

But okay, yeah, but if you're cheaper, VCs are like, eh, is that a good long term business model?

Pablo Srugo (34:05)

That's not cool. 

G(34:07)

It was actually very hard. Exactly, it’s not cool. They didn't have something that they could go tell all their partners and all their friends. They're like, oh, they got something really cool going on here, and so we actually were kind of bad at that. I would say it wasn't until our series C maybe that we actually got a pitch that kind of resonated and it ended up being a very technical pitch where we talked about complex distributed joins and I'm would draw architecture diagrams for these VCs to try to convince them of what's going on, and we got it to the point where it was crisp enough that people could kind of get behind it

Pablo Srugo (34:40):

And at some point your numbers are doing a big part of the talking.

Jeffrey Wang (34:44):

Honestly, even up until series C, the numbers were not enough. They really wondered, what's going on here? How can I qualitatively say that I'm excited to invest in this thing, and so it's like half BS story and half based on reality, a way of talking about what we're building that would kind of capture this idea of, hey, there's all these questions that people want to answer that they have a hard time answering in your traditional analytics tool or an out of the box data warehouse and we're really good at those. It was just a very nuanced thing to communicate that we had seen so many times with so many customers in practise, and so we just intrinsically knew it, but it was very hard to communicate, and I think that actually kind of hurt our marketing ability, our marketability or whichever one, it's like we couldn't figure out how to talk about it in the right way. We kind of tiptoed that for a long time, navigated all sorts of different paths.

Pablo Srugo (35:37):

I'm particularly interested because right now I am leading a pre-seed deal on a late entrant, like a company that is very similar to others and internally it was a bit of a battle even though it's so early. I know the founders, there's a lot of different things, but the reality is many people are stuck and maybe you're right, we'll see- TBD, but they believe that it's like, wow, this is the same. You know what I mean? There's already a market leader, why would you do this? And I'm like, but they're winning deals against this and that. So my question to you is now, and having obviously been able to do for so long and understand and you zoned out and you have hindsight 2020, if you could go back to let's say a series A pitch where you've got numbers, you've got traction, you've got proof points, but as you said, your story wasn't sharp, the differences were too nuanced. Is there something you think you could have said? Is there some story you think you could have told at that time to make it more obvious?

Jeffrey Wang (36:31):

This might come off as a little bit strange or taboo, but it's like I think you should try to almost make up a story that is rooted in reality to some extent, but exaggerates in some interesting ways to communicate your point, and that's a little bit where we ended up by the time the series he ran around, whereas I would say Spencer, Curtis and I we're very just honest people if not to brag about honesty, but we're kind of bad at spinning the story in our words in a way that oversells and so we should kind of say it like it is. I value a lot that a lot in our working culture, 

Pablo Srugo (37:10):

Well, even your marketing, if you think about that 10 million was based in reality, here's 10 million credits boom.

Jeffrey Wang (37:15):

Exactly, exactly. Yeah, maybe we should have done a hundred million if we're exaggerating. There's a little bit of, there are some domains, marketing, fundraising where I think you get disproportionate leverage out of your ability to story tell and exaggerate in ways that, not trying to lie, but trying to communicate your point again in a sharper and more emotional way to people so that they have something to latch onto and that's good for recruiting guess for marketing fundraising, it has a lot of benefits, and so it's like, I'm not looking down on this at all. I think this is actually great and a super valuable tool and skill for founders to have, and I think we're kind of bad at it,

Pablo Srugo (37:58):

So what would it be- like right now if you could go back and just say, okay, here's why we're different, what would that be? 

Jeffrey Wang (38:04):

Yeah. I think actually leaning on this kind of infrastructure piece of, Hey, we designed this crazy new database and it totally changes the cost profile as well as enables all these things. Again, we ended up kind of coming to that but seriously, we could have still told that from the beginning and there would be some truth in it, and then I think that just much more easily communicates what's going on.

Pablo Srugo (38:27):

Everybody wants that secret sauce. Sauce part is the story. A hundred percent,

Jeffrey Wang (38:31):

Man, when people used to ask me, what's your secret sauce? I'd be like, oh, we don't really have a secret sauce. That's a bad answer. Don't say that. Don't tell anybody that. Just say you have a secret sauce. It's like, yeah, we're the smartest engineers. We got all this experience building distributed systems. We came up with all these crazy algorithms to solve the problem. It's just say all of that and it's like there's actually some truth to all of it, but my personality was such that at the time I couldn't just say those things and take myself seriously

Pablo Srugo (38:59):

Funny. On the one hand, this is the world, and so if you want to fundraise, and obviously many companies need to or can benefit from fundraising, there's obviously a benefit of storytelling. Obviously. The other side of this is the VCs, the people in my shoes. It's like you look at big companies, we talk about Walmart, Walmart's secret sauce. I charge less. Literally, that was the stuff. Discount retailing is like people charge too much. I'm going to charge less and I'm going to put it in cheap places and more people are going to come, and then you have a hundred billion dollars plus company, Costco, what's your secret sauce? I buy less stuff and more of it, and that's it. Home Depot, it's weird, but massive businesses get built on the back of little subtleties fundamentally and obviously just incredible entrepreneurs. I think that that's what I always go back to, but then in cases like this too, you do, you have to look at the reality of it. Like, okay, there's no clean VC story, but customers are choosing them above the other ones. Maybe that is the bigger thread to pull on.

Jeffrey Wang (40:03):

yeah, it is a lot of- I kind of understand. It's a lot of work. If you evaluate a lot of deals, going to talk to a bunch of customers. It's just a lot of work, and so by the time we got further along, then you're doing fewer deals at the series B and C stages, and so then they spend a bunch of time with customers, so the combination of the revenue, the customer stories, the pitch, it all comes together a little bit more easily. Even having gone through the experience that we went through, if I were to evaluate myself as a prospective investor, I'm not sure that I actually do a great job of it, and so I don't want to say that this just this insight that this is possible and necessarily makes it easy to evaluate, and a lot of it does come down to obviously stuff like luck, but also just the founders and their- as an example, one of the things that I believe that allowed us to succeed is we just worked really freaking hard. It's hard to communicate that. It's like, oh, what'd you do? We just worked really hard. It's like, it's like, do I believe

Pablo Srugo (41:05):

Hard? How hard? Yeah, especially those early days when there's five, you, 10, 20 of you, how many days a week do you work? How many hours?

Jeffrey Wang (41:12):

A bunch of us lived just across the street from the office, and so we were just in there every day from-

Pablo Srugo (41:17):

Saturday, Sunday,

Jeffrey Wang (41:18):

Yeah, yeah, Saturday, Sunday, 10:00 AM to 10:00 PM That was our life, and we were pretty transparent about that with people that we were recruiting and things like that, and sometimes maybe unnecessarily transparent. They'd be like, oh. yeah, I don't want to do that, and we lost some good candidates because of that, but at the same time, it's like if that's how we actually are, you don't want to bring people on.

Pablo Srugo (41:42):

You might as well be upfront and until when you were doing this in 2014, do you think you that

Jeffrey Wang (41:47):

15 for the most part, again, once you got to 20 people, it's not like everybody's doing that exactly, but for a core set of people

Pablo Srugo (41:54):

but the founders set the bar. If the core founders are doing it, then everybody measures themselves to that. Maybe they don't do that, but that sets the bar on what does being all in really mean.

Jeffrey Wang (42:04):

Exactly. Yeah. I want to say through 2016, wow. We were in there all the time, and yeah, one it was, I don't even want to paint it as like, oh, we were just on this insane grind and just burning ourselves up. I actually had a lot of fun. It was really fun in those years for a couple reasons. One, it's like just the people, my co-founder, Spencer and Curtis, but also all the people we hired were just very high energy invested in making this thing happen, believed that we were an awesome team that could work together and accomplish these things, and part of me, I reminisce a little bit about the pre-covid days where everybody was in the office and you could actually feel that energy every hour. It's like, yeah, we are making progress on a really interesting and hard problem from a building perspective, from a selling perspective, from a company culture defining perspective. We all kind of believed in this in a way that I think it's very hard to replicate those types of environments, and that's the thing that sustains you. You burn out not because you're just spending too many hours on something, you burn out because you don't have that sort of energy sustaining environment that allows you to go beyond what you might expect yourself to be able to do,

Pablo Srugo (43:23):

And in your case, you guys, you were winning in the sense that you were seeing the results, the fruits of your labour, so to speak. 

Jeffrey Wang (43:29):

Yes. Winning. Winning is an important part of it. Yeah. Yeah. I won't deny that. It's much harder to stay motivated if things aren't working. That said, there were ups and downs, lots of ups and downs. We had some brutal quarters, we had some brutal outages.

Pablo Srugo (43:43):

When you think about your downs, your punches in the face, what are some of the ones you really remember?

Jeffrey Wang (43:48):

Q1 2016 is the one that is an unforgettable one, but, we came off an incredible 2015, so that was good.

Pablo Srugo (43:56):

Where were you at now? Yeah, walk me through those because you did what a million in 2014, what was kind of your ARR?

Jeffrey Wang (44:00):

A million ARR in 2014? At the end of 2015, we were at four and a half or so, a really, really good 2015, and so we were coming off this great high. We came back from the winter break and New Year's and everything, and then on January 4th we accidentally ran a script that deleted a bunch of critical databases.

Speaker 2 (44:22):

Oh man,

Jeffrey Wang (44:23):

And this was at 10:00 PM. We were in the office. Someone accidentally ran a script and I was like, oh, shit. It's like, wait, what do we do? It's like this thing's gone. Everything, all the alarms are firing because the whole site's down. Our whole data collection pipeline is down because there's these critical databases and we're like, oh, shit, it's, we're panicking. We're calling up AWS. It's like, Hey, can you recover this? We accidentally deleted these three databases and they're like, oh, well, it's not a normal thing. We can't actually recover the databases, but we can give you all these S3 files that are dumps of the underlying data structures that were behind the database, and we're like, okay, whatever. We'll take it. We'll work with it. So that was the longest I've gone without sleeping ever in my life, which is three days is not insane,

Pablo Srugo (45:14):

But still, it's a serious amount.

Jeffrey Wang (45:16):

We basically just worked nonstop to bring it back, and we eventually did, but it was brutal. We had three days where most of the product wasn't working, and then another week after where we basically had to catch up. At some point we got the data collection back up, but we couldn't process it. We were just accumulating this huge, massive data, multiple days of data, of a backlog, and it took us another week after to actually catch up, so it was like 10 plus days until Amplitude was working properly again. Absolutely brutal. There were points during that where I was like, oh, shit, this company might be gone. This is such a customer trust eroding incident, and if we're not able to recover the data, then it's like,

Pablo Srugo (45:59):

Yeah, then it's done. Then it's absolutely done.

Jeffrey Wang (46:00):

Pretty hopeless. At that point

Pablo Srugo (46:01):

did you have any churn or, I mean, 10 days is probably not enough, I'm assuming most waited.

Jeffrey Wang (46:06):

What ended up happening is, one, we were incredibly transparent to our customers. A lot of them were small at the time, so they were taking a bet on us, and so we were like, Hey, we totally messed up. This is really bad. It can't happen again. We will refund you some percentage of your contract and all of that. We actually ended up with no churn, which was incredible on the part of our customer success team at the time. In particular, the relationship that they'd built, the communication that they came up with to really, we sent a very, very heartfelt apology to our customers, and I think they responded well to that and so much appreciation to anyone who was an Amplitude customer at the time. Those were some tough times for us, and it must've been a pain in the ass for them to be like Hey, why is this product not working at all? I Just bought it. We were trying to raise our series B at the time as well, and oh my God, not talk about the incident, but it surfaced from a few customers that the VCs talked to, and so we kind of worked around it at the same time. -

Pablo Srugo (47:06):

they must have been like “Don't you guys have the most advanced data structure ever?” <laughs>

Jeffrey Wang (47:11):

Both of those things at the same time? Oh, yeah. Just like, Hey, we're the best engineers. That quarter was also brutal from a sales perspective. We were not doing well. I don't remember what the reason was exactly. It was a slow quarter. It happens every once in a while, and we came up with all sorts of silly things. We had thing, which we called it the Leap Year special, which just for enterprise sales company

Pablo Srugo (47:34):

Have discounts to just find- 

Jeffrey Wang (47:37):

To just find customers, and then we lovingly refer to it now as the China bailout of Q1 2016. We actually signed, believe it or not, we signed ByteDance as a customer in 2016 for $20,000 a month, and they were sending, not doin, not the TikTok equivalent that was too big, but they had a couple of other apps that were quite large in China, and they became one of our largest customers at the time. Now, in true Chinese customer fashion, they eventually churn and build their own analytics thing, but classic that. Yeah. The last week, these random, it was actually ByteDance and another Chinese company. They came in and they gave us probably 350K of ARR between them and saved the quarter after the outage, after the fundraising troubles, and that was quite the quarter, and nothing will be the same as that, thankfully.

Pablo Srugo (48:35):

That's awesome. And then you finished that year actually quite strong, right?

Jeffrey Wang (48:38):

Yeah, yeah. Year ended up being great. We ended up over 13 a million in ARR, and so-

Pablo Srugo (48:43):

That's huge.

Jeffrey Wang (48:43):

Yeah, it all worked out. That quarter was rough, but then we picked it up from there. At the time it was, again, there was a lot of inbound from- still even then from Mixpanel, people who were on big Mixpanel contracts and looking to the next thing, and over those first couple of years, honestly, a lot of it was just like X customer would be like, Hey, we like you more the Mixpanel, but there's ABC features that we need, and then we'll sign. It's like, okay, great. We'll build those. That was a response every single time. We'll just build those features, whatever they are. And that was our roadmap for the first three years of the product. Pretty straightforward when you put it that way. And that's how we kind of approached building the company and the market up.

Pablo Srugo (49:30):

Oh, that's all. my summary of it from the outside looking in. And I don't mean to simplify, but it really, you guys kind of kept things simple in a way. I mean, obviously there's so much harder work and I think innovation in terms of the structure, but it's like, listen, we're going to do what they're kind of doing, but we're going to do it a little bit better and certainly a lot cheaper, and then we're just going to do what customers want and really stay. That whole story around Q1 2016, like you said, I think there's the craziness of it, but I think what shines out to me is the relationships, the customer service that you must have had underlying everything to not have any churn. You can't just turn that on when the problem happens. You have to have before 

Jeffrey Wang (50:05):

Yeah, most Definitely

Pablo Srugo (50:05):

and you haven't slowed down. I think myself 13 and a half, 2016 now, eight years later. I mean, in 2020, I guess you crossed a hundred million, and now you're doing about 300 million.

Jeffrey Wang (50:17):

We're doing over 300 now,

Pablo Srugo (50:18):

12 months. And ARR looking at it, it must be well over 300. So Wow. It's a pretty crazy journey, man.

Jeffrey Wang (50:24):

Yeah. I cannot deny. It has been crazy.

Pablo Srugo (50:27):

So let's stop it there. Let me ask the two questions we always end on. The first one is for you, when did you feel like you guys hit product market fit?

Jeffrey Wang (50:35):

Yeah, good question. I would say in 2014, if I were to think back, there were two deals that year that really stand out to me as one that was surprising to us, which usually is a sign of something going on, and they were really important for that year. One was actually Nokia, and so Nokia, they used to make cell phones and whatnot, but now they primarily, or at the time, they primarily were making a mapping app called here. It was kind of like a GPS navigation type app. And for whatever reason, that team was kind of very much at the forefront of modern startup technology. And so they bought a lot of tools like us, but that was the first deal that we would consider enterprise in the traditional sense. It wasn't that big of a deal. I think they were paying $3,000 a month, but when we closed that, we were like, oh, wow, this established huge logo, huge company. Yeah, well, I could tell my parents about this company, that kind of thing. It's like they're actually buying us, and they loved using Amplitude, and the guy that bought us there ended up being- He is still a champion for us today, like 11 years later at many other companies. And so it's like that was a really, really important proof point of, Hey, we can sell to enterprise companies.

Pablo Srugo (51:54):

It's funny, I just think what comes to mind is like, this is long enough. It wasn't 10 years ago, but it's long enough ago that Nokia was probably a bigger name than ByteDance to vast majority of people.

Jeffrey Wang (52:05):

Oh, nobody knew ByteDance? We didn't know what ByteDance was. Chinese company is coming crazy. They have a lot of volume. It's kind of interesting. Yeah, Nokia was much bigger than ByteDance back then. And this was because it was 2014. It actually was. Yeah, it was 11 years ago almost now. So that was a big deal that year. And then the second deal that really stands out was Coursera, who also closed in 2014. They paid us. They were the biggest deal at the time. They paid us $10,000 a month. And so that was probably close to half or a third of our ARR at the time, because it was a summer 2014 that was like, wow, this company's willing to pay us a lot of money for this-

Pablo Srugo (52:49):

over 100k yeah.

Jeffrey Wang (52:49):

That's big. Almost an unimaginable amount of money at the time. And it was like, okay. I mean, that tells us something really interesting. What we're building is actually valuable. People value this thing a lot. They actually need this to do their jobs and be successful. And what we built actually solves that problem for them. And that was, I think the combination of those two was there is something real here for sure. And so we were fortunate that it didn't take too long to get to that point. Combination of the market being a little bit established and us having some good luck at the beginning with traction. I think those two really convinced me that, hey, there is product market fit here.

Pablo Srugo (53:30):

And then the last question is, taking everything you've learned over the last decade, what's one of the biggest pieces of advice you'd have for an early stage founder today?

Jeffrey Wang (53:40):

Yeah. Yeah. There's a lot of options. I think the one that I come back to, if I were to think about what I wish I told myself, it would be to have obsessively high standards about things. As a person, that's kind of what I gravitate toward naturally. But I think what I didn't quite realise that Spencer, our CEO, often reminds me of is that, Hey, we are building this company. It is our company. And that means you basically get to decide everything to an extent. And so if something is bad, one, it's your fault, it's your company, this is your fault. But two, it's like you can change it in almost arbitrary ways.

I think one of these things where most people in life, you're going through life and the vast majority of stuff happening around you, you kind of don't have control over. That's just how it's the big world, lots of things going on. And so you kind of get into this just normal habit. It's like, oh, this thing kind of sucks and I just have to deal with it. That's life. And that's actually healthy for the vast majority of things, because otherwise you'd probably be really stressed out all the time, but you just live in that way. But if it's your company, you don't need to think in that way. You can actually control everything. And not in, I'm a controlling micromanagement type of way, but you can infuse your own beliefs, your own convictions, your own principles into everything that goes on at the company. And then you want to bring people who believe in those as well, because otherwise you probably won't work very well with them.

(55:20):

But also you want to reinforce the things that you believe in. And because it's your company, you don't have the same obligation to just deal with things that just happen in the world. And I think leaning more into that would've- one: made me a little happier perhaps, because you get kind of caught up in like, oh, these bad things are happening, and it's like, I wish I could change that. It's like, no, you actually could change that and spend the time on it and think about it. I think maybe it's the other thing. Sometimes something bothers you and you don't really realise why, but often the case is there's something that you believe in that is not being reflected in something that's happening. And so taking the time to think about that and encode that, one thing I'm really happy with is we did that for our cultural values at the beginning of Amplitude. So humility, ownership, growth, and mindset are our values. And I'm so happy we took the time to think about that, write that down, and propagate that to this day without that, a bunch of other things that the company would've gone wrong. And I would've been like, Hey, why did that happen? Well, it's because we didn't really think about it. And so yeah, having really high standards about things that you care about, internalising that as one of the most useful levers for building a company, and then hopefully making yourself happy as a result.

Pablo Srugo (56:35):

Love that, man. Well, Jeff, thanks so much for jumping on the show, man. It's been great.

Jeffrey Wang (56:38):

Of course. Thank you. Pablo, it was a pleasure to have this conversation.

Pablo Srugo (56:42):

So picture this, it's months from now, years from now, and one of your founder friends, a really close founder, friends of yours, guess what? Their startup went bankrupt. And it turns out, if you had just shared the product market fit show with them, they would've learned everything they needed to find product, market fit, and to create a huge success. But instead, their startup has completely failed. You have blood on your hands. Don't let that happen. You don't want to live like that. It is terrible. So do what you need to do. Tell them about the show. Send it to them, put it on WhatsApp, put it on Slack, put it where you need to put it. Just make sure they know about it and they check it out.